mm/gup: Move page table entry dereference into helper function (0005d20b) · Commits · 戴 / test

This is a preparation patch for the transition of x86 to the generic GUP_fast() implementation. On x86 PAE, page table entry is larger than sizeof(long) and we would need to provide a helper that can read the entry atomically. Signed-off-by:

Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Aneesh Kumar K . V <aneesh.kumar@linux.vnet.ibm.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Dann Frazier <dann.frazier@canonical.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rik van Riel <riel@redhat.com> Cc: Steve Capper <steve.capper@linaro.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-arch@vger.kernel.org Cc: linux-mm@kvack.org Link: http://lkml.kernel.org/r/20170316152655.37789-4-kirill.shutemov@linux.intel.com Signed-off-by:

Ingo Molnar <mingo@kernel.org>

mm/gup.c

+12 −8

Original line number	Original line	Diff line number	Diff line
	@@ -1189,6 +1189,17 @@ struct page *get_dump_page(unsigned long addr)
	*/		*/
	#ifdef CONFIG_HAVE_GENERIC_RCU_GUP		#ifdef CONFIG_HAVE_GENERIC_RCU_GUP

			#ifndef gup_get_pte
			/*
			* We assume that the PTE can be read atomically. If this is not the case for
			* your architecture, please provide the helper.
			*/
			static inline pte_t gup_get_pte(pte_t *ptep)
			{
			return READ_ONCE(*ptep);
			}
			#endif

	#ifdef __HAVE_ARCH_PTE_SPECIAL		#ifdef __HAVE_ARCH_PTE_SPECIAL
	static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,		static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
	int write, struct page *pages, int nr)		int write, struct page *pages, int nr)
	@@ -1198,14 +1209,7 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,

	ptem = ptep = pte_offset_map(&pmd, addr);		ptem = ptep = pte_offset_map(&pmd, addr);
	do {		do {
	/*		pte_t pte = gup_get_pte(ptep);
	* In the line below we are assuming that the pte can be read
	* atomically. If this is not the case for your architecture,
	* please wrap this in a helper function!
	*
	* for an example see gup_get_pte in arch/x86/mm/gup.c
	*/
	pte_t pte = READ_ONCE(*ptep);
	struct page head, page;		struct page head, page;

	/*		/*

Admin message